There is a pressing public health need to detect Alzheimer's disease (AD) in its earliest, asymptomatic, stages. This would immensely facilitate the conduct of targeted clinical trials, the development of disease-modifying therapeutic agents, and, ultimately, the curtailment of the looming epidemic that AD poses. Such an endeavor necessarily requires a multidisciplinary team of investigators with complementary areas of expertise. The goal of this Beeson Patient-Oriented Research Career Development Award in Aging (K23) proposal is to provide the candidate with the experience, knowledge, and skillset necessary to carry out high quality, clinically- relevant, aging research so that he might effectively participate in and lead such a multidisciplinary group in the future. The proposal, therefore, comprises a unified set of research and training activities that are well-tuned to the candidate's transition from a K23 trainee to an independent investigator. The training plan seamlessly combines meetings with mentors, formal coursework, didactic activities, hands-on training, leadership training, and professional development. Specific training goals include: (1) cultivate a more nuanced understanding of aging and geriatric cognitive disorders, (2) receive dedicated training in neuroimaging methods, (3) develop advanced neuroimaging data analytic expertise, (4) receive ongoing training in the responsible conduct of research, and (5) obtain the career guidance needed for successful transition from the K23 to an independent research career, and maturation into a future leader in the neuroimaging of preclinical AD. In turn, the overall objectives of the research project are to use novel multi-modality machine learning techniques to specify the characteristic pattern of brain changes that is distinctive of persons in Stage 3 preclinical AD (the stage hypothesized to impart the greatest risk of future progression to AD), and then determine whether asymptomatic, middle-aged adults who exhibit such brain changes are more likely to experience future cognitive decline. These objectives are directly relevant to the global effort to halt AD via early detection of cognitively-healthy persons at high risk for progressing to AD. This is because current national guidelines for detecting risk for AD i asymptomatic persons call for extensive evaluations (e.g., nuclear imaging and lumbar puncture) that are expensive, not always well-tolerated by research volunteers and, more importantly, not widely available. In contrast, this project will rigorously assess brain changes i Stage 3 preclinical AD using routine, non-invasive, and broadly-deployable magnetic resonance imaging (MRI) measurements of brain structure and blood flow. A specific deliverable of this project is the derivation of a single, quantitative, abnormality score that can be used-in combination with pertinent health information-for identifying, on an individual level, asymptomatic persons at heightened risk for AD. Such persons may then benefit from more extensive AD biomarker testing, closer monitoring, and treatment with disease-modifying drugs when such drugs become available. The project's success has material potential to significantly extend the public health reach of the proposed guidelines for defining preclinical AD. The three specific aims addressed in this project are: (1) specify the pattern of brain changes on MRI that is characteristic of Stage 3 preclinical AD, (2) prospectively assess the prognostic utility of an aggregate measure of midlife structural-functional MRI brain changes, and (3) preliminarily evaluate how individual differences related to cognitive reserve and genetic risk modify the association between early brain changes and future decline. Completion of the interrelated set of research and training activities proposed in this K23 will greatly foster the candidate's development into an independent clinician-scientist with expertise in conducting translational and multidisciplinary neuroimaging studies of preclinical AD.